Advanced baseball statistical analysis offers fans the opportunity to better understand their favorite teams and players, highlight drivers of success, suss out underlying problem areas, and generally illuminate the mechanics of the game.

Sometimes, though, despite an array of positive individual indicators, it can feel like a team’s overall results are not amounting to jack squat. Lately, this is the sensation experienced by many fans of the Detroit Tigers, a team that only recently ended its worst losing streak in ten years and finds itself stuck in third place in the AL Central, 3.5 games back of both Minnesota and Kansas City. Miguel Cabrera’s swinging the bat well enough by his own standard to disguise the fact that he probably still isn’t 100% healthy. Ian Kinsler’s getting on base. Glove-first shortstop Jose Iglesias still is in the top ten in batting average. Glove-first center fielder Anthony Gose, former owner of a plus-.300 average, has not faded badly, sustaining a .289/.328/.399 line, good for 103 wRC+. In other words, the Tigers were hitting, they just weren’t winning, and it was hard to place the bulk of the blame anywhere other than on the offense. This conundrum underlined the rather fundamental point that there is a difference between getting on base and scoring, and the former is only interesting to the extent it acts as a useful proxy for the latter.

To visualize this, I plotted every team’s run opportunities (i.e., baserunners), calculated as the sum of their hits, walks, and hit by pitches, against actual runs scored in an attempt to show the degree to which teams were able to convert baserunners into runs:

on_bases (3)

(Data through June 7, 2015 from Baseball Reference. Thanks to fellow ALDLAND contributor physguy for graphical assistance.)


The above provides evidentiary support for the probably unsurprising conclusion that getting on base is reasonably meaningfully correlated with scoring runs. It also shows that Detroit and San Francisco have put the most men aboard, but both have been relatively ineffective at converting those baserunners into runs. Toronto’s offense is going gangbusters, but their outlier plot, together with their sub-.500 record, serves as a reminder that defense matters too– the only AL team to allow more runs is Boston. On the lower end of the spectrum, we are reminded that few teams are doing more with less right now than Minnesota, and no team is doing less with less than Philadelphia.

That helpless feeling Tigers fans had? It’s real, and it’s shared by Giants fans. The 2012 World Series foes, with their matching 4.12 runs per game, have similar records — 30-28 for Detroit, 32-26 for San Francisco. It makes sense that hitting into double plays would tamp down run-conversion efficacy, and the Tigers have a stranglehold on that dubious title. Although the numbers are more tightly distributed, the Tigers also find themselves near the top of the caught-stealing leaderboard, where they currently sit in second place. (Detroit has radically transformed its base-stealing strategy over the past few seasons, so this could be a symptom of growing pains in that regard.) The Giants, meanwhile, are only slightly above average on both counts, but they are stranding more of their own baserunners than any other team by a healthy margin, For San Francisco, strikeouts don’t appear to be the culprit, but their batted-ball profile seems to provide the answer: their ground-ball percentage is the fourth-highest in baseball and their ground ball to fly ball ratio is third-highest. This isn’t a problem on its own, but it is when coupled with an above-average soft-contact percentage. The result? One of the worst infield hit percentages in baseball. In other words, the Giants are hitting a lot of soft grounders that defenses are easily turning into outs.

The above graph and subsequent speculation are attempts to understand what has happened so far this year. With two-thirds of the 2015 season remaining, I suspect that teams will find themselves moving closer to that regression line for the same reason Pythagorean-based projections tend to be generally accurate. I also suspect that sequencing — the way teams cluster their hitting and other getting-on-base actions– explains a not-insignificant portion of teams’ deviations from the charted regression line, and that, at least early in the season, a team might reasonably prefer the current position occupied by Detroit and San Francisco to the one occupied by Minnesota. Assuming something like luck explains why those teams find themselves where they are now, in the long run of a full season, better inputs (more baserunners) should somewhat naturally yield better outputs (more runs). Play on.

More of AD’s work may be found at ALDLAND.

Next post:
Previous post:

3 Responses to “Visualizing Run Conversion Efficacy”

  1. Peter Denton

    Nice post. I think that this metric can be used to describe luck. Amount above or below the best fit line is how lucky or unlucky a team is in the context of sequencing. Batting average (and the related advanced versions of it) seem to be a fairly truly random variable. That means that, as the season goes on, the relative (not absolute) spread in the figure should decrease. I think it is correct to say that the teams on the bottom have been unlucky so far and, while they will probably end the season on the bottom half, each game that goes by is just as likely to move them up as down, so the teams on the right side should score more runs, on average, even if they haven’t been so far.



  1.  The 2016 Detroit Tigers in a Not-Quite-Six-Feet-Under Box | Banished to the Pen

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.